Mobile devices have become the most widely used camera platforms. However, acquiring real-time images of the paper surface during planar operations such as writing and drawing remains challenging. Positioning a camera above the paper surface enables straightforward imaging but blocks the user’s line of sight and restricts head and arm motion. Capturing images from the side of documents avoids these issues but introduces trapezoidal distortion and reduces resolution in the distal region of the paper. This study presents a modular side-view objective that is integrated with a tablet camera and omits the sensor and autofocus module, enabling cost-effective imaging in large-scale production. Based on a theoretical analysis of pupil aberrations, we introduce field angle constraints and controlled entrance pupil size errors to minimize trapezoidal distortion and enhance the resolution in the distal region. The proposed side-view objective employs a freeform mirror to generate a relay pupil conjugate to the entrance pupil with reduced pupil aberrations, thereby suppressing stray light. To match the exit pupil to the tablet camera, an iterative design method evaluates and corrects exit pupil aberrations using light-footprint geometry, yielding an accurate exit pupil. Performance and tolerance analyses demonstrate that the design achieves a modulation transfer function above 0.25 at 100 cycles/mm, effectively controls trapezoidal distortion, suppresses stray light, remains compatible with tablet cameras, and maintains good manufacturability.

Abstract Object detection, a relatively recent development, allows for the identification of multiple objects in photographs. Numerous machine learning algorithms have been developed for object detection, particularly for small‐scale objects. Small‐object detection algorithms can be used to count a wide variety of objects from counts of individuals in drone imagery to cell counts on a microscopic slide. These algorithms include faster region‐based convolutional neural network (R‐CNN), You Only Look Once (YOLO) and single shot detector (SSD). Each has its own strengths and weaknesses. To measure immune response in animal populations to both biotic (e.g. disease exposure) and abiotic conditions (e.g. temperature), researchers rely on manual counting of cells, which is time‐intensive. For insects, this research centres on counting haemocytes. For controlling insect pests, species‐specific viruses often represent a major biocontrol agent. However, we still lack knowledge about how the immune system reacts to being challenged with a pathogen via haemocyte counts. Here, we use machine learning, which automatically counts haemocytes, to compare detection algorithms. To perform counts, we trained Faster R‐CNN, YOLO and SSD models on 398 photographs and validated training on 114 photographs. Of the algorithms compared, YOLOv8 was the most accurate and the quickest to train. Counts performed on YOLOv8 tended to be fairly accurate and matched closely with haemocyte counts done by hand. YOLOv5 followed closely behind in accuracy with SSD and Faster‐RCNN not far behind. Counting time was reduced from 60 s per photograph when counting by hand to less than a second using YOLOv8's detection capabilities. Practical Implication . Object detection methods such as these drastically speed up the process of measuring immune response and open the door for further research whether it be on biocontrol methods in insects or other questions of biological interest. In general, small object detection allows for quick processing of otherwise difficult data and can apply to multiple fields including use in drone imagery, counting cells and detecting crop yield from photographs.

This article presents the project Observing South Caucasus’ Historical Landscape: An Open Photo Archive , whose strategic objectives are to collect, digitise, and catalogue a collection of photographs documenting the cultural heritage of the South Caucasus. The preservation and valorisation actions are targeted at three different facets of this heritage: (1) the tangible photographic collection (i.e. the photographic object itself); (2) the intangible historical layers (evolution of site ownership, historical stratification, and corresponding toponymy); (3) the tangible architectural and natural heritage of historical Armenia and Georgia.

Abstract I first read Fredric Jameson not long after I graduated from college in 1977. His Marxism and Form (1971) was my introduction to Lukács, Bloch, Benjamin, and Adorno, and most of the book flew over my head. One passage stuck with me, though; it is an early example of how effectively Jameson uses specific practices to track general shifts in the phenomenology of capitalist culture: We need only juxtapose the mannequin, as a symbol, with the photographic objects of pop art, the Campbell's soup can [and] the pictures of Marilyn Monroe … we need only exchange, for that environment of small workshops and store counters, for the marché aux puces and the stalls in the streets, the gasoline stations along American superhighways, the glossy photographs in the magazines, or the cellophane paradise of an American drugstore, in order to realize that the objects of Surrealism are gone without a trace. Henceforth, in … postindustrial capitalism, the products with which we are furnished are utterly without depth…. All libidinal investment in such objects is precluded from the outset.1

Historiographical work, like all sciences, is a constant appeal to all accumulated knowledge, to evidence placed in the light of criticism of the categories and concepts that order the world of knowledge. In this second decade of the 21st century, photography has become a protagonist, given the amount of information brought by the arts and by all the digital manipulation of visual and sound images. We have divided the text into three parts: first, the photographic object, memory and truth; then, photography as art; then, photography as a source of research and our final considerations.

The paper presents a study of the effectiveness of using deep learning models U-Net, DeepLabV3+ and Feature Pyramid Network (FPN) for semantic segmentation of aerial images. This task allows to automate the analysis of large amounts of data in various areas of human activity (urban planning, environmental monitoring, etc.). In the process of the study, our own input data was prepared and models were trained. To evaluate the experimental work, the loss function, accuracy, Jaccard index, F-measure, and prediction time of one image were calculated, which allowed us to compare the quality of segmentation. The results show that each model is able to effectively recognize the main objects in aerial photographs. The U-Net model demonstrates better speed during training, DeepLabV3+ trains longer, but shows better performance, even on a small amount of input data, and FPN provides the best balance between speed and quality at the same time. The results can be used for further research in the field of automated image analysis. Figs.: 5. Tbls.: 3. Refs.: 13 titles.

Pictotrans is a revolutionary multilingual platform that transforms language translation and cultural comprehension by the integration of advanced image recognition and real-time translation technologies. The platform enables users to photograph objects and receive precise translations in their desired language, supplemented with culturally sensitive insights and context-aware descriptions. The platform is intended for travellers, language learners, and those who are involved in cross-cultural communication, with both functional application and educational purpose. In contrast to traditional translation software, Pictotrans places a strong focus on context and cultural meaning in addition to its translation feature. Users can capture photos of objects to receive translated text in their chosen language, complemented by cultural insights and contextual phrases. By overcoming the shortcomings of tools as Google Lens and Microsoft Translator, Pictotrans is an essential solution. Its capacity to seamlessly integrate multilingual translation, cultural insights, and advanced image recognition makes it a perfect travel companion for travellers in pursuit of meaningful experiences and language learners in pursuit of intensive language acquisition. With its user-friendly interface and contextual focus, Pictotrans transforms the way people interact with languages and cultures, breaking barriers and encouraging world connections.

With thousands of observation satellites orbiting the Earth, massive-scale ultrahigh-definition (UHD) images are captured daily, covering vast areas of land, often extending across millions of square kilometers. These images commonly feature a wide range of ground objects, such as vehicles and rooftops, numbering from tens to hundreds. The ability to categorize the diverse types of objects in UHD aerial photographs is essential for a variety of real-world applications, including intelligent transportation systems, disaster prediction, and precision agriculture. In this study, we introduce a novel framework for categorizing UHD aerial photographs. The core of our approach is to represent the spatial configurations of ground objects topologically and encode these layouts using a binary matrix factorization (MF) technique that robustly addresses the challenge of noisy image-level labels. Specifically, for each UHD aerial photograph, we identify visually and semantically important object patches. These patches are then connected spatially to form graphlets, small graphs that capture the layout and relations between adjacent objects. To enhance the understanding of these graphlets, we propose a binary MF approach that captures their semantic content. The method integrates four key components: 1) learning binary hash codes; 2) refining noisy labels; 3) incorporating deep image-level semantics; and 4) adaptively updating the data graph. The binary MF is solved iteratively, with each graphlet being transformed into a set of discrete hash codes. These hash codes, which represent the spatial and semantic information of the graphlets, are subsequently encoded into a feature vector using a kernel machine, enabling multilabel categorization of the aerial photographs. For validation, we compiled a large-scale dataset of UHD aerial photographs, sourced from 100 of the top-ranked cities worldwide. Experimental results demonstrate that: 1) our method excels in learning categorization models from imperfect labels and 2) the integration of the four proposed attributes enables effective encoding of the graphlets into hash codes, providing a powerful representation of the UHD aerial photographs.

Until now, achieving both a high spectral resolution on the order of a few wavenumbers and the highest sensitivity in Raman scattering spectroscopy has required reliance on high-end laboratory instruments. Here, we introduce an innovative yet design-wise simple alternative: a cost-effective and compact micro-Raman spectrometer (µRS) that combines exceptional spectral resolution and sensitivity. Leveraging industrial-grade CMOS cameras and high-quality photographic objectives, our µRS maintains a footprint at least five times smaller than traditional lab-based spectrometers. Through detailed characterization and direct experimental comparison, which includes the use of calcite as a Raman standard, we demonstrate that our µRS achieves a spectral resolution of down to 2.5 cm−1. Using a single-layer MoS2 sample, we found that the sensitivity of our system, while somewhat lower, remains within a useful range compared to commercial research-grade confocal Raman microscopy systems. This study presents a compelling solution for researchers seeking efficient and high-resolution Raman spectroscopy tools across diverse applications, particularly in resource-limited or field-based settings.

The automatic detection of objects in medical photographs is an essential component of the diagnostic procedure. The issue of early-stage brain tumor detection has progressed significantly with the use of deep learning algorithms (DLA), particularly convolutional neural networks (CNN). The issue lies in the fact that these algorithms necessitate a training phase involving a large database over several hundred images, which can be time-consuming and require complex computational infrastructure. This study aimed to comprehensively evaluate a proposed method, which relies on an active contour algorithm, for identifying and distinguishing brain tumors in magnetic resonance images. We tested the proposed algorithm using 50 brain images, specifically focusing on glioma tumors, while 2000 images were used for DLA from the BRATS Challenges 2021. The proposed segmentation method is made up of an active contour algorithm, an anisotropic diffusion filter for pre-processing, active contour segmentation (Chan-Vese), and morphologic operations for segmentation refinement. We evaluated its performance using various metrics, such as accuracy, precision, sensitivity, specificity, Jaccard index, Dice index, and Hausdorff distance. The proposed method provided an average of the first six performance metrics of 0.96, which is higher than most classical image segmentation methods and was comparable to the deep learning methods, which have an average performance score of 0.98. These results indicate its ability to detect brain tumors accurately and rapidly. The results section provided both numerical and visual insights into the similarity between segmented and ground truth tumor areas. The findings of this study highlighted the potential of computer-based methods in improving brain tumor identification using magnetic resonance imaging. Future work must validate the efficacy of these segmentation approaches across different brain tumor categories and improve computing efficiency to integrate the technology into clinical processes.

The use of deep-learning algorithms in the production of photographs has challenged a key distinction within media images: that between the photograph and the frame. Theory has often emphasized the non-coincidence of these two modes of the photographic, one incomplete and lacking its own visibility, the other autonomous, unique, and finite. But today, the intervention of algorithms at the stage of image capture weakens this distinction. Even the single photographic snapshot becomes a “vertical” concentration of several images, while an apparently fixed photograph can conceal a “horizontal” series of images that also allow the picture to be viewed as a clip. In the face of these “dense” snapshots, which no longer correspond to human perception and imply a very different dialectic between trace and visualization, singular and plural, snapshot and series, does it still make sense to juxtapose the photographic object and the photogrammatic material used within numerous types of imaging? Or should the photographic be redefined as a genetically serial process?

ABSTRACT This article analyzes NoPhoto Collective’s Memoria colonizada photography projects (2011–2018), which feature some of the new agricultural colonies established by the Francoist regime between the 1940s and the 1960s through its Instituto Nacional de Colonización (INC). The stated aim of these projects is to reflect upon “the intervened landscape, the creation of new populations with a uniform architecture, official memory and that of the protagonists, giving special importance to the local actors” (NoPhoto Collective. 2011. Vegaviana memoria colonizada. NOPHOTO. Accessed 28 June 2024. http://nophoto.org/vegaviana-memoria-colonizada;http://vegaviana.nophoto.org/). Yet, the foregrounding of personal and intimate images and stories, while intended as a counterpoint to the villages’ official histories and a means to empower local residents, risks presenting a nostalgic and somewhat idealized view of these communities. Such slippages into idealization and cliché, however, coexist with photographs that mobilize representational strategies that undermine any illusion of photographic objectivity or neutrality, and which thus throw into question both the supposed transparency of the photographic medium and the notion of a collective rural memory that articulates the series as a whole.

At the end of the nineteenth century, the advent of x-ray machines fueled American medicine's reliance on technology, transforming hospitals and the medical profession. X-ray manufacturers pursued the nascent hospital market as competition and patent feuds accelerated x-ray machine modifications. Hospitals incorporated clunky new machines and employed x-ray photographers, but as the unruly apparatus stabilized, physicians joining the new specialty of radiology discounted the toils of machine troubleshooting and promoted their medically qualified x-ray interpretations. This article frames early medical radiography in terms of boundary work, highlighting how discourse among physicians, x-ray photographers, and hospital administrators vied to establish a privileged demarcation between radiological science and photographic craft. Ultimately, radiologists supplanted x-ray photographers by leveraging the automation of x-ray machines and capitalizing on the epistemic shift from photographic objectivity to qualified interpretations. By focusing on this overlooked aspect of x-ray incorporation into hospitals, this work provides a unique perspective on how harnessing mechanization and authoritative medical interpretations can shift professional boundaries.

This article examines the visualization of Hutsuls in German-Austrian, Ukrainian, Polish and Russophile ethnographic texts, asking how national and imperial imaginations of space were produced through such fluid cross-linking of texts and photographs. Considering the radical changes in image circulation since the late-19th century, we aim to reconsider the role of photography in image-making of the Habsburg Empire. This article shows how the same images were supposed to serve many purposes, when they were embedded in different settings. The construction of photographic objectivity, the circulation of images through imperial infrastructures and the exoticization of rural peoples were, however, common phenomena.

Background A neural network is a machine learning algorithm that can learn and make predictions by adjusting the strength of the connections between nodes. The sigmoid function is commonly used as an activation function in these nodes. This study explores the potential applicability of biological materials in the development of alternative activation functions. Methods Inspired by the fact that acetylcholine induces intestinal contractions that follow a sigmoid function, we used pharmacological data obtained from guinea pig ilea in a layered neural network for image classification tasks. Results and Conclusions We found that the intestinal data-based neural network with the same structure as a conventional three-layer perceptron achieved an impressive classification accuracy of 85.7% ± 0.6% based on the MNIST handwritten digit dataset (chance = 10%). Additionally, the neural network was trained to determine whether objects in photographs collected from the internet were digestible, achieving an accuracy of 88.5% ± 0.9% (chance = 50%). Our approach highlights the potential applicability of intestine data in neural computations based on pharmacological mechanisms.

The traditional food of Lima Puluh Kota Regency as a form of traditional culinary culture needs to be re-actualized, namely by displaying it through visual images that are more attractive, refreshing and have a contemporary nuance through a work of food photography. In the culinary industry, food photography is an important requirement, both for commercial and non-commercial purposes. This is because artistic visual images can increase the image and selling value of the food displayed. In realizing this, several methods of creating work were carried out, namely qualitative descriptive methods with observation activities, literature studies and interviews. The data obtained was then analyzed qualitatively and presented descriptively. Then, the design method and work realization were carried out to realize the concept that had been designed, with the shooting process carried out indoors using a Cannon EOS 1200D camera and other supporting equipment. The creation of the work "Traditional Food of Lima Puluh Kota Regency in Food Photography" includes the food Sate Danguang-Danguang, Unja Ikan, Kipang Ompiang, and Galamai Boreh Rondang as photographic objects. From this food, 4 photo works were created and it is hoped that the creators can overcome the problem of preserving the traditional food of Lima Puluh Kota Regency, becoming one of the superior regional culinary and cultural products, and can increase the interest of local people and tourists in consuming traditional food, especially in the Lima Puluh Kota Regency.

The differences in albumen photographs from vintage photographic studios were identified by energy-dispersive X-ray fluorescence spectroscopy and Fourier transform infrared spectroscopy. The results inspired the concept of finding common features characteristic of a given photographic studio. The obtained measurement data (i.e., positions of vibrational bands for characteristic groups of albumen and the mass contents of chosen elements) were analyzed chemometrically by employing the Principal Component Analysis (PCA). The PCA technique allowed us to reduce the number of relevant experimental parameters characterizing the unique features of the photographic objects. The two major components were able to distinguish the photographic objects in terms of their authorship and the time to produce a photograph. The method developed was examined for a selected group of photographs consisting of albumen prints from three Polish photographic ateliers. To validate ED-XRF measurements and, consequently, the chemometric findings, reference albumen photo samples were designed and prepared. The empirical functional relationships between the content of photochemically reduced silver particles on the photographic paper and several physicochemical factors, including time of exposure to UV light, AgNO3 concentration in a fixed bath, and concentrations of other additives, were proposed. These results can be used for the prediction of the experimental conditions under which the investigated photographs were developed.

This paper is related to the study of the features of the neural technologies’ application, in particular, ResNet neural networks for the classification of objects in photographs. The work aims to increase the accuracy of recognition and classification of objects in photographs by using various models of the ResNet neural network. The paper analyzes the features of the application of the corresponding models in comparison with other architectures of deep neural networks and evaluates their efficiency and accuracy in the classification of objects in photographs. The process of data formation for training neural networks, their processing and sorting is described. A web application and a mobile application for recognizing and classifying objects in a photo were also developed. A system for classifying objects, in particular airplanes in photographs, was developed using neural network technologies. It gives a recognition and classification accuracy of about 95%. Research results of ResNet models are of great practical importance, as they can improve the classification accuracy of various images. Features of ResNet, such as the use of skip connections or residual connections, make it effective in the relevant tasks. The results of the study will help to implement ResNet in various fields, including medicine, automatic pattern recognition and other areas where the classification of objects in photographs is an important task.

Given a composite image with photographic object and painterly background, painterly image harmonization targets at stylizing the composite object to be compatible with the background. Despite the competitive performance of existing painterly harmonization works, they did not fully leverage the painterly objects in artistic paintings. In this work, we explore learning from painterly objects for painterly image harmonization. In particular, we learn a mapping from background style and object information to object style based on painterly objects in artistic paintings. With the learnt mapping, we can hallucinate the target style of composite object, which is used to harmonize encoder feature maps to produce the harmonized image. Extensive experiments on the benchmark dataset demonstrate the effectiveness of our proposed method.

Painterly image harmonization aims to harmonize a photographic foreground object on the painterly background. Different from previous auto-encoder based harmonization networks, we develop a progressive multi-stage harmonization network, which harmonizes the composite foreground from low-level styles (e.g., color, simple texture) to high-level styles (e.g., complex texture). Our network has better interpretability and harmonization performance. Moreover, we design an early-exit strategy to automatically decide the proper stage to exit, which can skip the unnecessary and even harmful late stages. Extensive experiments on the benchmark dataset demonstrate the effectiveness of our progressive harmonization network.